Janus AI: Abu Dhabi researchers develop a visual search engine that guides surgeons during robotic operations
A research team from Mohamed bin Zayed University of Artificial Intelligence in Abu Dhabi won an international award in Italy after developing an artificial intelligence tool called “Janus AI”, designed to provide immediate visual guidance to doctors during robotic surgeries by retrieving matching clips from previous successful operations.
The new system acts as a specialized search engine for robotic surgery clips, analyzing the live video feed from inside the operating room and automatically retrieving a 30-second video segment from a prior surgery that matches precisely the stage the surgeon is currently at. It gives doctors the ability to compare up to three previous surgical cases within fractions of a second with a single button press, helping the attending or evaluating surgeon explore how other colleagues handled the same complex step and transition safely to the next stage.
The name was inspired by the Roman god Janus, who has two faces, one looking to the past and the other to the future, accurately reflecting the system’s mechanism that relies on an archive of previous surgical experiences to guide real-time medical decisions. The innovation won the Innovation Challenge in Robotic Urology Surgery at the Eros 26 conference hosted by the Italian city of Padova last September, after a competition that brought together international computing and surgery communities.
Professor Shezari Stefanini, a professor of robotics at the university, oversaw the technology’s development, working with post-doctoral researcher Eduardo Vazari who led the software development effort for about five months this year following extensive discussions with clinical practitioners. The project also involved doctors and surgeons from the University of Illinois at Chicago and the University of Turin in Italy, with the current visual database focusing on robotic urological procedures, especially kidney removal, and the team looking to expand the library to other robotic surgeries.
The system is not yet ready for use in live surgeries.It requires additional refinement stages and the attainment of regulatory and clinical approvals before it can be adopted in hospitals, while the researchers aim to forge a commercial partnership to launch the product on the market once development is complete, with plans to attract new graduate students, including Emirati researchers.
This achievement has a direct impact on health-sector plans in the Gulf and Egypt, where major hospitals have poured massive investments into acquiring advanced robotic surgery systems while still facing a gap in training young local staff and preparing them to independently conduct precise interventions. Converting disabled archival recordings stored on hospital servers into an interactive visual reference shortens the learning curve for resident doctors, and gives hospitals and health investors an opportunity to reduce surgical error rates and redirect external consultancy budgets toward completing compliance frameworks and clinical accreditation for locally developed medical software.